Does Mass Deworming Affect Child Nutrition? : Meta-Analysis, Cost-Effectiveness, and Statistical Power

Croke, Kevin ; Hicks, Joan Hamory ; Hsu, Eric ; Kremer, Michael ; Miguel, Edward

World Bank, Washington, DC

关键词: child nutrition;    deworming;    meta-analysis;    early childhood development;    cost-benefit analysis;   

DOI  :  10.1596/1813-9450-7921
RP-ID  :  WPS7921

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

The WHO has recently debated whether toreaffirm its long-standing recommendation of mass drugadministration (MDA) in areas with more than 20 percentprevalence of soil-transmitted helminths (hookworm,whipworm, and roundworm). There is consensus that therelevant deworming drugs are safe and effective, so the keyquestion facing policymakers is whether the expectedbenefits of MDA exceed the roughly $0.30 per treatment cost.The literature on long run educational and economic impactsof deworming suggests that this is the case. However, arecent meta-analysis by Taylor-Robinson et al. (2015),(hereafter TMSDG), disputes these findings. The authorsconclude that while treatment of children known to beinfected increases weight by 0.75 kg (95 percent CI: 0.24,1.26; p=0.0038), there is substantial evidence that MDA hasno impact on weight or other child outcomes. This paperupdates the TMSDG analysis by including studies omitted fromthat analysis and extracting additional data from includedstudies, and finds that the TMSDG analysis is underpowered:Power is inadequate to rule out weight gain effects thatwould make MDA cost effective relative to comparableinterventions in similar populations, and underpowered toreject the hypothesis that the effect of MDA is differentfrom the effect that might expected, given deworming'seffects on those known to be infected. The hypothesis of acommon zero effect of multiple-dose MDA deworming on childweight at longest follow-up is rejected at the 10 percentlevel using the TMSDG dataset, and with a p value < 0.001using the updated sample. In the full sample, includingstudies in settings where prevalence is low enough that theWHO does not recommend deworming, the average effect onchild weight is 0.134 kg (95 percent CI: 0.031, 0.236,random effects). In environments with greater than 20percent prevalence, where the WHO recommends mass treatment,the average effect on child weight is 0.148 kg (95 percentCI: 0.039, 0.258). The implied average effect of MDA oninfected children in the full sample is 0.301 kg. At 0.22 kgper U.S. dollar, the estimated average weight gain perdollar is more than 35 times that from school feedingprograms as estimated in RCTs. Under-powered meta-analysesare common in health research, and this methodological issuewill be increasingly important as growing numbers ofeconomists and other social scientists conduct meta-analysis.